Yeah, yeah, so I'm a little interested in the effect of the process of care and how it applies to patients who have CRTs in the post-op period. And so the case that I have is really asking questions of you, so we talk a little bit about how our practice is and how we go about evaluating and then changing things for patients. So let's start with this 78-year-old lady. So she has paroxysmal atrial fibrillation. She has mild to moderate MR on her echo. She is on a sololid rivaroxaban, and she comes in with new onset heart failure. She's in atrial fibrillation. She has a rapid ventricular response. Her BNP's increased. She's got crackles. I don't have to tell you about what a heart failure presentation is like. And now re-evaluation after admission and diuresis shows she's got a marked reduction in left ventricular function. So she went from an EF of 60 now to 27. She's sick in heart failure. The catheter in it's negative. She's diagnosed with idiopathic dilated cardiomyopathy. She has no other reversible potential contributors to her heart failure. So she goes on Amio. She gets diuresed. She gets a ramipril with a dose titrated up. She gets a low dose of bisoprolol. That combination of bisoprolol and Amio is being given to provide a combination of rate control initially and then contribute to rate control assuming you can get her out of heart failure. So a month later she sees the heart function clinic and she has spironolactone added. Three months later she changes from ramipril to antresto. The heart failure doc orders an echo and refers the patient to you or to the EP clinic for evaluation regarding the merits of a device intervention. So here's her ECG when she was in sinus rhythm. And what you see here, just to point out a few things, one is EPs we think about. So the first is she's in sinus rhythm. That's good news. Second thing is she has PVCs. She has wide complex beats that are probably best explained as PVCs. And she has a very wide QRS with an atypical left spinal branch block. And some would say that she doesn't quite fit criteria for a spinal branch block because she has an IVCD. But nonetheless she certainly has a wide QRS. So she then comes into your office as an EP and she brings her son with her. He's a software engineer, knows lots about these things, has done a bunch of research and so on. And the question is she's still in class two heart failure. She's got a wide QRS. She's got non ischemic dilated cardiomyopathy. She seems like a good CRT candidate from the standpoint of being a responder. Maybe on the borderline of being an ICD candidate because a combination of her EF and Danish like data suggesting she's in a low risk situation. You back and forth with this and decide to put in a CRTD. So you're now the first follow up for this lady who then underwent her CRTD implantation. The comments from the implant procedure was that it was quite unremarkable. But she comes for her one month follow up and she says, you know, I thought this thing was going to make me feel better. That's kind of what I signed up for. And I don't feel any better. I'm still short of breath at times and every once in a while, especially at night, I get hiccups and that's new. This is her symptom status. So I intentionally did not choose a vendor to make this a vendor platform question around programmer or programming type of thing. I intentionally wanted to say this is care process and there's some lots of similarities across the vendor platforms around what the issues are that you're checking for in a CRTD follow up. So she was programmed to pace from the distal pair in a quadripolar LV lead. The output 3.5, which is nominal on the device, the threshold was reported and you reproduce the threshold in the order of 1.3. All her other settings were left at nominal. So lower rate of 60, upper rate of 130 in this instance. Her counters indicate that she's paced by V91% of the time. Her AF and mode switching components indicate she's in AF27% of the time. The average heart rate, if you look at the rate histograms, is 91. And there's one episode of ATP. And when you look at it, your interpretation of that, without going into detail, looks like it's actually atrial fibrillation with a rapid ventricular response with ATP intervening. And her PBC count says it's about 2,000 a day. And ventricular sensor response, or the LV, if you like, triggered pacing, is also turned on. So I pose the following questions. And we can take these one by one, but I might ask the question, how do you test for phrenic nerve stimulation in a patient who volunteers she's having the odd hiccup? So can anyone tell me what they do? Yeah. How do you volunteer here? It was just in the email. OK. OK. So if you had trouble hearing it, he said, what we do is we turn up the juice and then see if we can capture phrenic. Yeah. My first question is, are they bothered by it? OK. Because there might be a reason that the vector is chosen. And if it's infrequent and it doesn't bother them, I'm less likely to ask. OK. If you didn't hear that, the point was focused around the question about essentially what the symptom burden is, whether it's something they notice or they voice. Yeah. So over here, and then Suneil. So if they can't, how do they recreate the position that they're at and see if we can test it? OK. And that's exactly my thought process. Well, I asked them about the circumstances. So this lady said she has hiccups at night. And then I asked, well, how do you sleep at night? And some people sleep supine. Some people love to sleep on their left side. And so I try and reproduce that same scenario. Yeah. So we're back to a point made earlier around the whole point about the historical context is very helpful. You can feel diaphragmatic infections. Pardon me? So put it this way. One of my points to make was that if this is something that's troubling to her, actually the very first thing I would do is actually go back and read the implant details to see if there's anything about this. So the comment about was this the sole vector? Why is the lead where it is? Where is the lead? And so on. And so, for instance, when I do an implant, if I find PNS is an issue, I put that in my letter. I want people to know we looked around, and the only good place we could find had a threshold of two and a diaphragmatic threshold of five. So that if they say they have the odd hiccup, which is going to happen sometimes, and it's hard to get around that necessarily at implant, I want the follow-up person to know there's a risk of phrenic nerve stimulation here, and because, for instance, two of the branches were unacceptable or couldn't be cannulated or something like that. So I think one of the key points in this situation has to do with, if you like, the symptom burden and the context for this. And I think the other point is what I think nominal is that we do most often is we turn up the juice and see if you can reproduce it, and you can both feel it and so on. But the other thing is that postural maneuvers can help bring it out. And that's something that's sometimes forgotten because the patient may not necessarily remember, for example, like they may say, I always sleep on my left side. And so then, for instance, I'll say, in what position do you wake up in? And the answer is almost never on your left side, because you've moved around in your sleep, and you wake up lying flat or on the other side. So it tells you that their sleeping position isn't necessarily consistent. Can I make one point that's relevant, but a little aside? We're not very good at asking about, many of us are not very good at asking about sleep patterns, especially sleep apnea. There are cases where sleep apnea really will subvert the whole issue of diaphragmatic pressure, your, my potential, et cetera. And we've got to start asking these patients about that. And in principle, their bed partner, right? Who is the most observant person, because usually that person is sleep or semi-asleep. Chuck, did you have a? So, yeah, I was just going to say, let's say there's a good reason we want to keep this pacing vector. Sometimes, as you know, since the cronax, you're right, for nerve stimulation is shorter than the cronax for myocardial stimulation. Sometimes you can program lower voltage outputs, right, with wider pulse widths. And get a little bit of, you know, increase your safety margin that way. Of course. Yeah. This is a perfect lead into my next question, which was, the device is programmed to pace from the distal pair on a quadrupolar lead. And we know, essentially, the more apical, the more anterior, or even middle cardiac pain, probably the less likely the patients are to get a benefit. So the question is, when somebody does this with the pacing bipole, do you revisit which pairs you're pacing from? And so a good example might be, in dilated cardiomyopathy, it's not unusual that proximal thresholds are higher because of the fibrotic pattern. But if you're going to set the output to 3.5, and the distal pair, for instance, has a threshold of 1.3 in this instance, and the proximal pair has a threshold of 2, does it really matter? Do you guess what's going to be more clinically effective? Chuck, you were working in this direction. Do you change which bipole based on intuitive probability of response? Right. So I think there are a couple of things. The first, broadly, this patient brings up the question of, what is a CRT non-responder? And how do we deal with this? So if a patient, and I'm sure he's going to get to this, but when we think about CRT response, we think about clinical functional response. We think about objective response, say, by echo being the measurements we typically use of heart failure. So let's assume this patient is not just depressed, and that's why she's not responding. Let's say you get an echo, and the ejection fraction's not any better. Then you ask the question, well, what can we do about this? It's different. So one of the questions, particularly in a patient who doesn't have left bundle branch block to start with, or typical left bundle anyway, is to say, is this part of the heart that we're pacing late in the QRS? And I think looking at the, particularly in this instance, this patient has sinus rhythm, looking at the QLV on all four unipolar leads is something that I do in non-responders routinely. So if the QLV is 100 and, say, 40, 150 milliseconds on the proximal pole, but 90 milliseconds on the distal pole, then even if the proximal pole has a higher threshold, I'd be interested in pacing there. If they're all within 10 milliseconds of each other, I pick the one that has the lowest pacing threshold. Anybody else from the audience? Yeah. Just one minor point that comes up in my clinic every three or four years. Somebody has a CRT device. They seem to be complaining of hiccups. We are checking the LV lead in multiple vectors, and maybe eventually, after half an hour, maybe the next clinic visit, somebody thinks to crank up the output on the RV lead, and we realize it's the RV apically placed lead. That's triggering diaphragmatic pacing intermittently. So I've just, I've come across that a couple of times. Because it's not, because you have direct diaphragmatic muscle stimulation, or people have chest wall stimulation, as opposed to, you know, phrenic nerve capture. It's not phrenic nerve. It's diaphragmatic stimulation. Right, right. Yeah. Yeah. Well, my thought is that it's probably quite premature to judge on response in this particular patient at this point of time, because she came back after one month. She had some subjective problems, 91% of pacing. We did not see the ECG. Perhaps the counters overestimate effective CRT pacing. She had right bundle branch block, with perhaps left anterior hemiblock pattern on the ECG. So I think there's, we still need to see the ECG on pacing. We need to deal with the AFib episodes, percentage of biventricular pacing, and then we might judge on her response. Yeah. And we're short on time. So, you know, one of the reasons to choose this type of case is, you know, there's a bunch of different parameters that contribute to optimization of the CRT state. And it includes the degree of pacing, competing rhythms, including AFib and PVCs, the position, the outputs and programming and so on, and then medical optimization, and also how much time she has. So I don't think even if everything was working perfectly and there was nothing to fix, you would abandon the possibility that she will respond. It's not uncommon for a response to take more time, and she's probably also not medically optimized. But part of the principle is to go through that checklist of things that contribute to why we put in CRTs and how we know to make them work well. And that will get us to most of the things that we can do to optimize her, for example. If I, how are we doing for time? We've got five minutes. Okay. I was going to say, do you want to say something about AF management, which is, of course, you know, and its role in affecting resynchronization? Yeah. Yeah. So that's a very good point. So first of all, one of the questions I have for her is whether her symptoms are continuous or proxismal. So if she has a 27% AF burden, what she may be voicing is there are periods of time when she's short of breath when she's basically in AF. And that becomes a clear, if you like, primary target for managing her further, as opposed to the notion that even in sinus rhythm with AV synchrony and VV synchrony to the right parameter, she's still generally unwell. What's clear is with an average heart rate of 91, when from her rate histogram, right, she is certainly overdriving her lower rate. She is in AFib, and there's the potential there for that to be a major contributor. And the other thing is that conduction in that situation can then be competing with your LV pacing. And depending on your platform, it's quite possible to be under-sensing competing rhythms from either AF conduction or PVCs. Is that sort of where you're going? Yeah. Exactly. Yeah. So there actually have been times where we will halt her to look for that or try to get some further information to look at those competing rhythms. And then what you realize is that you're really only pacing in the LV 70% of the time, and you're really underachieving. The other thing is our rate control strategies are typically quite empiric. And so, for instance, if you told a non-device patient that their average heart rate was 91, the trials would have told us that's decent rate control. But that doesn't mean it's not symptomatic or that there aren't times when the rate control is imperfect. And so part of this decision-making process leads to the question about robust both rhythm control and rate control for somebody who has AF. Any comments or questions from the audience on the area of AF management? Because here, for example, the no-brainer thing to do if she's got heart failure and blood pressure room is to just increase her beta blocker, assuming she's on the maximum dose of amiodarone you'd want as a maintenance rhythm control drug. Just a question on that, Matthew Webber from Wellington. That's a question. Assume that you go down the rate control strategy and we abandon the rhythm. We don't know whether, because the next escalation in rate control might be AV node ablation. And we don't know whether AV node ablation in this situation might be better than suboptimal rate control or suboptimal delivery of CRT pacing. Yeah, it's a real dilemma. So, for instance, I didn't present to you, but I did tell you that you had moderate MR. You know, if her MR is worse than her left atrium's 55 or her volume's 50, the probability of maintaining sinus rhythm is probably relatively low, even with amiodarone. And then your hand is forced around invasive strategies to address it. On the other hand, if the MR is better with medical therapy, her EF is coming up. She's a CRT responder and the LA's not that big. She's a candidate for ongoing medical or even ablation as a way to manage in terms of rhythm control. I think we all, I certainly warn a patient like this that we could be heading for AV node ablation. Okay, thank you. We have one more minute, and maybe I can ask you, Andrew, what do you make of that 12-lead ECG pattern, number one? Number two, one month post-procedure when her injection fraction with all of its problems is going up. What is wrong with just waiting and doing a Q three to six months echo and proceed from there? So first principles is, you know, do the least necessary to try to create some positive direction for this lady. So for instance, medical optimization and time is completely reasonable and is in a way a platform. And so then the next question is, are there some simple things you can do with programming to optimize the degree of LV pacing? To me, that's the strategy until she's at least at three months. The question at six months is if she's a little worse and the echo's no better and her MR is now moderate to severe, how you would intervene? The reality is, and dilated cardiomyopathy in particular, the probability of her getting back to being nearly class one is quite good. She just needs time. And your heart failure team will tell you, you know, anyone who's trying to get to state-of-the-art medical therapy in six weeks is likely to cause harm to the patient. So it's a journey of medical optimization as well. So things like BNP values, the other thing is exclusion of other contributors like anemia, COPD, thyroid disease, and these are sort of sleep apnea. These are also obvious things that we sometimes forget because we're a little cardio tunnel. But those are the other elements of the sort of holistic approach to the patient. So you know, this is an ECG that's not particularly encouraging as a probable responder, except the fact that she's female with a mechanism of cardiomyopathy that makes her a good CRT candidate. So at my place we have this. Every Monday morning we meet and discuss all the CRT and high voltage referrals from the week before and approve them collectively. Part of the reason is because these two A2B indications are a common component of the referral base and we want to agree that this is a reasonable next step for these patients.

heart failure

evaluation

management

patient care

atrial fibrillation